Our studies seek to define genetic changes that drive progression of melanoma, the most aggressive and deadliest skin cancer. New technologies have yielded a wealth of data describing the genomes of cancers. These studies are retrospective. They offer a catalog of defects present in individual cancers but do not discriminate between tumor-promoting genomic changes and those changes that arose incidentally during tumor progression. Functional studies are necessary to draw this distinction, and in many cancers the list of candidate genomic changes can be quite large. To parse through changes in gene copy number variation we use cross-species comparisons and functional studies with a conserved zebrafish melanoma model. Since some of the genomic changes involved in melanoma progression are seen in other cancers, our studies may provide broader insights into cancer development. The genetic changes we define may be useful as diagnostic and prognostic indicators of disease as well as therapeutic targets for melanoma and potentially other types of cancer. The two areas of study in this project are: 1. Define gene copy number changes that are shared between human and zebrafish melanomas, and use functional screening to identify new melanoma genes. Mechanisms of melanoma formation are conserved between zebrafish and humans. By determining genomic alterations common to zebrafish and human melanomas we can uncover new genes and mechanisms involved in melanomagenesis. 2. Examine the BMP-family ligand GDF6 for its roles in melanoma progression and melanocyte development. The comparative studies above have already identified an excellent candidate melanoma gene, GDF6. This BMP-family ligand is recurrently amplified and overexpressed in human and zebrafish melanomas, and its expression pattern during embryogenesis suggests a role in regulation of melanocyte development. We will assess how GDF6, and BMP signaling in general, participate in melanoma formation, perhaps by regulating the differentiation and proliferation of melanocytes.